PROPOSEDPOD 05: Secure AI & Model Ethics
Algorithmic Parity Audit of Student Merit Screening Baselines
Objective & Abstract
Applying AIF360 disparate impact metrics to inspect gender and demographic balance in automated applicant evaluation heuristics.
Investigated Methodologies
- Disparate impact calculation
- SHAP feature importance auditing
- Equalized odds constraint testing
Datasets & Benchmarking Corpora
Anonymized Historical Screening Data (Synthetic)
All training and evaluation datasets undergo ethics audit and licensing compliance per AIAC Research Council standards.